Cargando…
Introspective analysis of convolutional neural networks for improving discrimination performance and feature visualisation
Deep neural networks have been widely explored and utilised as a useful tool for feature extraction in computer vision and machine learning. It is often observed that the last fully connected (FC) layers of convolutional neural network possess higher discrimination power as compared to the convoluti...
Autores principales: | , |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
PeerJ Inc.
2021
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8114803/ https://www.ncbi.nlm.nih.gov/pubmed/34013030 http://dx.doi.org/10.7717/peerj-cs.497 |
_version_ | 1783691120539074560 |
---|---|
author | Shafiq, Shakeel Azim, Tayyaba |
author_facet | Shafiq, Shakeel Azim, Tayyaba |
author_sort | Shafiq, Shakeel |
collection | PubMed |
description | Deep neural networks have been widely explored and utilised as a useful tool for feature extraction in computer vision and machine learning. It is often observed that the last fully connected (FC) layers of convolutional neural network possess higher discrimination power as compared to the convolutional and maxpooling layers whose goal is to preserve local and low-level information of the input image and down sample it to avoid overfitting. Inspired from the functionality of local binary pattern (LBP) operator, this paper proposes to induce discrimination into the mid layers of convolutional neural network by introducing a discriminatively boosted alternative to pooling (DBAP) layer that has shown to serve as a favourable replacement of early maxpooling layer in a convolutional neural network (CNN). A thorough research of the related works show that the proposed change in the neural architecture is novel and has not been proposed before to bring enhanced discrimination and feature visualisation power achieved from the mid layer features. The empirical results reveal that the introduction of DBAP layer in popular neural architectures such as AlexNet and LeNet produces competitive classification results in comparison to their baseline models as well as other ultra-deep models on several benchmark data sets. In addition, better visualisation of intermediate features can allow one to seek understanding and interpretation of black box behaviour of convolutional neural networks, used widely by the research community. |
format | Online Article Text |
id | pubmed-8114803 |
institution | National Center for Biotechnology Information |
language | English |
publishDate | 2021 |
publisher | PeerJ Inc. |
record_format | MEDLINE/PubMed |
spelling | pubmed-81148032021-05-18 Introspective analysis of convolutional neural networks for improving discrimination performance and feature visualisation Shafiq, Shakeel Azim, Tayyaba PeerJ Comput Sci Algorithms and Analysis of Algorithms Deep neural networks have been widely explored and utilised as a useful tool for feature extraction in computer vision and machine learning. It is often observed that the last fully connected (FC) layers of convolutional neural network possess higher discrimination power as compared to the convolutional and maxpooling layers whose goal is to preserve local and low-level information of the input image and down sample it to avoid overfitting. Inspired from the functionality of local binary pattern (LBP) operator, this paper proposes to induce discrimination into the mid layers of convolutional neural network by introducing a discriminatively boosted alternative to pooling (DBAP) layer that has shown to serve as a favourable replacement of early maxpooling layer in a convolutional neural network (CNN). A thorough research of the related works show that the proposed change in the neural architecture is novel and has not been proposed before to bring enhanced discrimination and feature visualisation power achieved from the mid layer features. The empirical results reveal that the introduction of DBAP layer in popular neural architectures such as AlexNet and LeNet produces competitive classification results in comparison to their baseline models as well as other ultra-deep models on several benchmark data sets. In addition, better visualisation of intermediate features can allow one to seek understanding and interpretation of black box behaviour of convolutional neural networks, used widely by the research community. PeerJ Inc. 2021-05-04 /pmc/articles/PMC8114803/ /pubmed/34013030 http://dx.doi.org/10.7717/peerj-cs.497 Text en © 2021 Shafiq and Azim https://creativecommons.org/licenses/by/4.0/This is an open access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ Computer Science) and either DOI or URL of the article must be cited. |
spellingShingle | Algorithms and Analysis of Algorithms Shafiq, Shakeel Azim, Tayyaba Introspective analysis of convolutional neural networks for improving discrimination performance and feature visualisation |
title | Introspective analysis of convolutional neural networks for improving discrimination performance and feature visualisation |
title_full | Introspective analysis of convolutional neural networks for improving discrimination performance and feature visualisation |
title_fullStr | Introspective analysis of convolutional neural networks for improving discrimination performance and feature visualisation |
title_full_unstemmed | Introspective analysis of convolutional neural networks for improving discrimination performance and feature visualisation |
title_short | Introspective analysis of convolutional neural networks for improving discrimination performance and feature visualisation |
title_sort | introspective analysis of convolutional neural networks for improving discrimination performance and feature visualisation |
topic | Algorithms and Analysis of Algorithms |
url | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8114803/ https://www.ncbi.nlm.nih.gov/pubmed/34013030 http://dx.doi.org/10.7717/peerj-cs.497 |
work_keys_str_mv | AT shafiqshakeel introspectiveanalysisofconvolutionalneuralnetworksforimprovingdiscriminationperformanceandfeaturevisualisation AT azimtayyaba introspectiveanalysisofconvolutionalneuralnetworksforimprovingdiscriminationperformanceandfeaturevisualisation |